Blogging has an additional advantage of freedom. You can write whatever you want, and you are the judge here. In a community blog, you need to go through a scrutiny, and the admin needs to decide whether the content is appropriate or not for the community. From that perspective, blogging is much more flexible.
I am in the middle of the PhD study, very important time for a PhD student. This is the time you need to select your point of concerns, the research topic. I have been working on this for the last 1 year 8 months. By this time, I got some quality publications in the top venues like ICSE, SANER. But, for some reasons, it seems to me that my research topic is not fixed yet. I need to select a potential topic that will not die out within the next 10 years. I really feel sorry for my supervisor. That guy worked on clone detection and analysis, and the area is pretty much dying out. From several conferences and industry panels, what I have learned that people do not much care about the existence of clones. They are OK with clones which might create problems occasionally. But the way the researchers warned and shouted, the real scenarios are completely different. This is quite an embarrassment, and it is hard to be a part of it. So, I am looking for a thriving area for pursuing my research career.
For the last 1.5 years, I spent time mostly after Crowdsourced knowledge. It is a kind of knowledge which is accumulated by large technical crowd in various online repositories-- programming Q & A sites/forums, open source repositories, and various discussion boards. Specially, my target is Stack Overflow, the largest programming Q & A site, hosting about 10 million questions and serving about 4 million users. This is a large body of knowledge, specially programming related knowledge. My goal is to harness that knowledge effectively for the betterment of software engineering activities. During MSc, I mostly tried to incorporate such knowledge into the IDE, and my focus was on IDE-based search and recommendation. In PhD, my focus is on more sophisticated mining of various artifacts from Stack Overflow, and derive meaningful artifacts. So far, I have been successful in doing that. For example, we were able to provide insight for code comprehension, code search and code debugging. Now, I am trying to exploit the crowd knowledge in concept location query reformulation. So, I am pretty much into it. Since my BSc background was information retrieval, my PhD thesis will be a blending on crowdsourced knowledge, information retrieval, software maintenance, recommendation system and machine learning.
Yeah, the task is not simple, but I am into it. Hopefully, this will come out as a success, and I will get some more good quality publications. Recently I am trying for an ASE paper, and already submitted two papers in FSE and ICSME. Lets see, what happens. Until then, lets hope for the best. A second read could improve the writing, but who has the time?
I am in the middle of the PhD study, very important time for a PhD student. This is the time you need to select your point of concerns, the research topic. I have been working on this for the last 1 year 8 months. By this time, I got some quality publications in the top venues like ICSE, SANER. But, for some reasons, it seems to me that my research topic is not fixed yet. I need to select a potential topic that will not die out within the next 10 years. I really feel sorry for my supervisor. That guy worked on clone detection and analysis, and the area is pretty much dying out. From several conferences and industry panels, what I have learned that people do not much care about the existence of clones. They are OK with clones which might create problems occasionally. But the way the researchers warned and shouted, the real scenarios are completely different. This is quite an embarrassment, and it is hard to be a part of it. So, I am looking for a thriving area for pursuing my research career.
For the last 1.5 years, I spent time mostly after Crowdsourced knowledge. It is a kind of knowledge which is accumulated by large technical crowd in various online repositories-- programming Q & A sites/forums, open source repositories, and various discussion boards. Specially, my target is Stack Overflow, the largest programming Q & A site, hosting about 10 million questions and serving about 4 million users. This is a large body of knowledge, specially programming related knowledge. My goal is to harness that knowledge effectively for the betterment of software engineering activities. During MSc, I mostly tried to incorporate such knowledge into the IDE, and my focus was on IDE-based search and recommendation. In PhD, my focus is on more sophisticated mining of various artifacts from Stack Overflow, and derive meaningful artifacts. So far, I have been successful in doing that. For example, we were able to provide insight for code comprehension, code search and code debugging. Now, I am trying to exploit the crowd knowledge in concept location query reformulation. So, I am pretty much into it. Since my BSc background was information retrieval, my PhD thesis will be a blending on crowdsourced knowledge, information retrieval, software maintenance, recommendation system and machine learning.
Yeah, the task is not simple, but I am into it. Hopefully, this will come out as a success, and I will get some more good quality publications. Recently I am trying for an ASE paper, and already submitted two papers in FSE and ICSME. Lets see, what happens. Until then, lets hope for the best. A second read could improve the writing, but who has the time?
No comments:
Post a Comment